Document Type Master's Dissertation Author Bezuidenhout, Hannelie URN etd-09252008-142958 Document Title An electroglottographic and acoustic analysis of glottal activity during speech initiation in stuttering Degree MCommunication Pathology Department Communication Pathology Supervisor
Advisor Name Title Prof A van der Merwe Committee Chair Mrs E Groenewald Committee Co-Chair Keywords
- electroglottographic analysis
- physiological vocal fold activity
- acoustically detected vocal fold activity
- acoustic analysis
- speech sensorimotor control
- motor programming
- speech initiation
- glottal activity
Date 2007-04-24 Availability restricted AbstractStuttering is recognized as a motor speech disorder with aberrant glottal activity being a salient characteristic. Current models of stuttering view this disorder as a dysfunction in the neuromechanism with involvement of specific neural structures, such as the basal ganglia. Within a framework of speech sensorimotor control as proposed by Van der Merwe (1997), the characteristics of glottal activity in stuttering seem to be associated with a motor programming disorder. Results of previous electroglottographic (EGG) and acoustic studies on vocal fold activity in people who stutter (PWS) showed abnormalities in the rate, range and muscle tone of vocal fold activities and difficulties in voice initiation. These results support the notion of motor programming involvement in PWS. It is therefore proposed that a description of vocal fold activity during a speech task may contribute to a better understanding of speech motor control in stuttering.
The purpose of this study was to determine differences in vocal fold activity in PWS and normal fluent speakers (NFS) during speech initiation. An EGG and acoustic analysis was conducted to investigate physiological and acoustically detected vocal fold activity in the two groups. The effect of different phonetic contexts on vocal fold activity was considered during the compilation of the speech material, namely: voiced without articulation (vowel), voiced with articulation (C(v)), and voiceless with articulation (C(vl)). The subject group comprised six PWS with stuttering severity levels ranging between very mild to severe. Six NFS were matched with the PWS with regard to gender and age. These six participants served as control group. Each participant orally read five wordlists consisting of 12 VC, C(v) VC and C(vl) VC utterances. The EGG and acoustic recordings were made simultaneously. Only perceptually fluent speech samples were analyzed.
Results regarding physiological vocal fold activity showed the following: (a) Prephonatory vocal fold activity in PWS, compared to NFS, was mainly characterized by irregularities for VC, C(v)VC and C(vl) VC productions. (b) PWS had longer and more variable laryngeal onset time (LxOT) values for VC, C(v)VC and C(vl)VC utterances compared to NFS. (c) Motor complexity of the phonetic context affected the frequency of occurrence of disruptions in prephonatory vocal fold activity as well as the LxOT in both groups. For PWS, C(v)VC and C(vl)VC productions were more involved than VC productions.
The following results were obtained with regard to acoustically detected vocal fold activity: (a) Similar to NFS, the C(v)VC and C(vl)VC utterances of PWS were characterized by gradual voice onsets; VC productions showed a greater number of abrupt voice onsets; gradual voice onsets for PWS were associated with high frequencies of vocal fold irregularities in the period prior to voice onset. (b) Boths groups showed mean voice onset time (VOT) for initial voiced /b/ productions and initial voiceless /p/ productions within normal limits. (c) VOT range for /b/ productions was greater than for /p/ productions in NFS; this tendency was also seen in most PWS. (d) For PWS, VOT for /b/ productions were slower and overlapped with VOT for /p/ productions. The VOT range for /p/ productions showed consistency and a tendency toward a restricted VOT range. (e) As opposed to the NFS, VOT range values for /b/ and /p/ productions overlapped for most PWS; thus, VOT values for /b/ and /p/ productions did not indicate voice-voiceless contrasts for PWS.
The current findings on vocal fold activity during speech initiation in PWS were discussed within the framework of speech sensorimotor control with the emphasis on motor programming. From this perspective, deviant vocal fold activity in PWS may involve abnormal increase of muscle tone in the laryngeal structures or abnormaliteits in rate, velocity, and range of movements. Difficulties in the initiation of successive movements during the prephonatory period in PWS may also be associated with inefficient motor programming. The proposal that stuttering is a motor programming disorder could guide further explanations of speech motor control of PWS.
Copyright 2006, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.
Please cite as follows:
Bezuidenhout, H 2006, An electroglottographic and acoustic analysis of glottal activity during speech initiation in stuttering, MCommunication Pathology dissertation, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-09252008-142958 / >
© University of Pretoria 2006
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access dissertation.pdf 2.34 Mb 00:10:48 00:05:33 00:04:51 00:02:25 00:00:12indicates that a file or directory is accessible from the campus network only.